Mostly we are using U.S.-based data. There are a number of U.S. population-based surveys that have excellent data on sexual orientation, and more recently, on gender identity as well. There's the behavioural risk factor survey in the United States. The national survey of family growth is another one that has very good data. There are a number of surveys there that we tend to use to answer the questions that we're unable to answer with Canadian data.
On February 28th, 2019. See this statement in context.